Age-related susceptibility to disease is the most common cause of morbidity, mortality, and diminished quality of life. Although likely related to both genetics and epigenetics, the epigenetic influences on age- related disease have not been defined previously or related to genetic variation. We have recently proposed a novel paradigm for understanding the relationship between variation in DNA sequence, epigenetic marks, and phenotype that considers the role of genes and epigenetics in affecting not only the mean of a phenotype but also its plasticity, or variance. We have begun to apply our approach by identifying highly variably methylated regions (VMRs) in the human genome. We have discovered that VMRs define an epigenetic signature in aged individuals, and that they are associated with important health indicators such as BMI. The purpose of this proposal is to fully explore this new paradigm by identifying VMRs associated with common age-related phenotypes across 3 domains (body composition; energy availability/demand; maintenance of body homeostasis) and determine the relationship between VMRs and underlying genetic variation. In addition to the traditional view that genotypes directly control expected (or mean) phenotype values, we argue that another major effect of genotype is to control stochastic epigenetic variation leading to increased variability of methylation at a particular genomic site and to a consequent increased phenotypic variation. This is a dramatically new concept. We postulate a new kind of SNP, which we call a vSNP, or variation-SNP, that is associated with the degree of methylation variability at VMRs, rather than mean methylation level, and thus also associates with the spread, or variability, of a phenotype in a population rather than the mean value. Such a vSNP would increase the proportion of individuals at both high and low risk of disease, and thus would not be detectible in traditional association analyses. To support our approach, we show preliminary data identifying vSNPs related to VMRs, vSNPs related to increased phenotypic variance, and VMRs related to phenotype. Our aims are: 1 - To investigate the relationship between DNA sequence, DNA methylation, and quantitative aging phenotypes under both our mean and plasticity models, using 2000 participants in the Baltimore Longitudinal Study of Aging (BLSA), and genome-wide SNP and methylation approaches. 2 - To perform focused DNA sequencing and capture bisulfite sequencing to identify the specific sequence and epigenetic variants responsible for findings in Aim 1. This work will help elucidate how genetic variation relates to methylation variation and how both impact age-associated phenotypes that increase disease susceptibility. DNA methylation could be a new target for risk assessment and intervention that can reduce the burden of disease and disability and slow down the deleterious effects of aging.